SABRe CVD Project 1 Completion of discovery Px measurements of 135 MI cases and controls In-depth discovery Px was conducted on 135 myocardial infarction (MI) samples as a sub-set of the overall CVD sample set. Overall, the data were quite good with over 600 unique proteins being measured and over 350 proteins found in all 270 samples. The quality of the data is excellent for such a large study and statistical analysis is ongoing. Univariate and multivariate analysis of MI discovery Px data set The data were interrogated using univariate and multivariate strategies. Prior to the multivariate analysis a complex imputation methodology was designed to bolster the data set and increase the statistical power of the multivariate models. Results of these analyses were generally not publishable as a stand-alone unit, but the data were further utilized in 2 different paths. First, this data helped decide which proteins would be measured in the entire CVD data set using targeted MRM strategies, and second the data were used to provide protein targets for the SABRe CVD Project 2. Completion of protein measurements by targeted MRM analysis Using the aforementioned MI discovery Px data set, two panels of targets were identified for MRM analysis. The first panel is a direct-MRM measurement panel containing 33 target proteins. The second panel for MRM contains 38 targets which are lower concentration proteins which will require a depletion step prior to MRM assay execution. The direct MRM analysis has completed data collection and QC and the data on the full 676 sample set from the CVD study are currently undergoing statistical analysis. Completion of GCMS measurements on Metabolic Syndrome (MetSyn) Risk Factor study The entire Metabolic Syndrome risk factor study (652 samples) underwent GCMS analysis at a 3rd party facility. The QC of the data was completed demonstrating very good results. Overall there were >140 analytes measured with the 80% of these having a CV of <15% across all control samplesan excellent result. These data have recently begun statistical analysis and though very early in the process, the results are showing clear ability to differentiate amongst the different MetSyn risk factors. SABRe CVD Project 2 Completed assay development of three panels of target proteins The assay development lab was able to assess the overall list of 49 targets and quickly design and validate assays for 31 of these targets because of familiarity or an existing assay already in place. These 31 were binned into three appropriate conc. panels and further tested to meet the standards of multiplexed assays. Panel-1 (5 members) are high conc. targets with a plasma concentration of >5000ug/mL. Panel-2 (12 members) are med level concentration targets with a plasma conc. of >50ug/mL and finally Panel-3 (16 members) are level level conc. targets with a plasma conc. of <10ug/mL. Completion of measurement of targets against 7400 plasma samples Panel-1 with only 5 targets was tested first in the high-throughput assay system to better grasp the complexity of running 7400 samples. Technical challenges were overcome and Panel-1 was completed in eight weeks. Panel-2 was initiated in late August and is anticipated to be completed by the end of September 2011. Once Panel-2 completes, both Panel-1 and Panel-2 data will be exchanged and statistical analysis can be initiated. On-going assay development of 18 protein targets Of the 49 nominated targets, 31 have passed the assay development stage and 18 are active. These 18 sit at different steps in the development pathway, but most are nearing completion and have been assembled into three additional multiplexed panels. A few targets are undergoing reagent development since there was no appropriate commercial source of reagents, thus the lab is constructing new reagents to meet the multiplex assay requirements. SABRe CVD Project 3 Completion of chipping for the FHS Offspring study The data collection of gene expression analysis of FHS Offspring was completed in March of 2011. There were several technical set-backs in FY2011 the lab had to overcome, but other than losing ground against our timelines, the lab was up to the challenge and methodically worked through each issue as it arose and was able to successfully complete this unprecedented study of expression analysis on over 2600 samples. Data cleaning procedures defined and tested Completing this enormous study was one measure of success, but with this success come the next challenge of cleaning and conducting QC on this enormous data set. This time it was the CIT analysis team that rose to the occasion. Over a period of months, the team analyzed, modeled, corrected and re-analyzed until the absolute best set of technical variables were determined to create the absolute best data set possible from this wealth of data. These tremendous efforts will be rapidly implemented on the future FHS Third Generation samples set which is even larger than the Offspring cohort, as about 3400 participant samples from the Third Gen will be analyzed. CVD case-control study analysis 5 batches representing the 221 matched pairs in a CVD case-control sample set were processed in the lab and the gene expression data were generated. Overall, the data were good and analyses were conducted at the gene and exon level. These data have now been analyzed by many models and techniques with the best genes having been nominated for a validation study to confirm the results and the overall qunatitation of the array data. Data collection of the FHS Third Generation study samples The lab shifted to begin collection the 2nd part of the overall data set;this being the 3400 RNA samples from the Third Gen cohort. Thought about three months behind our original timeline (due to technical and instrument complications) the Third Gen data set is now making excellent progress. Over 22 batches have been completed (1900 samples) and the lab is working towards completing all Third Generation samples by the end of December 2011 (only a few months behind the Sept 2011 original timelinea remarkable accomplishment!) Data deposition of the case-control data to dbGaP In July 2011 the first SABRe CVD data set was deposited into dbGaP. This was the gene expression analysis case-control study of 442 samples. There are other future deposits to be made of the full Offspring data set in early 2012. SABRe CVD Project 4 Completions of the assay re-design and pilot study testing The new TaqMan miRNA assay was extensively tested in both non-Framingham and eventually Framingham samples to fully validate this approach after the failure of the SybrGreen assay in FY2010. The new QC results are outstanding and the amplification effects, batch effects and panel effects are minimal across all of the pilot and QC studies. Completion of the 476 CVD case-control measurements The lab team conducted miRNA analysis on the CVD case-control sample set that was defined for the Project 3 expression analysis. This set was completed by the lab prior to their relocation in April and the data passed along to the statistical team for QC and analysis. Overall, as expected the QC was excellent for this small data set and the data are presently undergoing statistical analysis. Identification of non-detectible miRNA The statistical team has taken the data from the case-control study and subsequently removed all miRNA that gave no signal from the RNA isolated from the Paxgene tubes. Though there are over 800 miRNA available for testing, we found that there only 400 of these that are detectible in the Paxgene RNA. The non-detectible RNA have been removed from the assay panels going forward, so there will be less regent cost and faster throughput for the laboratory.